commit | db45b6a33d3c8bae05e4544fc22ed4a6220537da | [log] [tgz] |
---|---|---|
author | Frank Barchard <fbarchard@google.com> | Wed Oct 09 16:42:45 2019 -0700 |
committer | XNNPACK Team <xnnpack-github-robot@google.com> | Wed Oct 09 16:45:23 2019 -0700 |
tree | cabc98916cc486cf1019b12bfb6b30dc80ab70ea | |
parent | 174706ec89e07015bdde31aac2c5574d9ad48c79 [diff] |
1x8 neonfma IGEMM microkernel and 1x8 benchmarks. Add new 1x8 neonfma IGEMM intrinsics microkernel. Add benchmarks for 1x8 neon and neonfma IGEMM microkernels. BUG=142398150,140592595 PiperOrigin-RevId: 273849950
XNNPACK is a highly optimized library of floating-point neural network inference operators for ARM, WebAssembly, and x86 (SSE2 level) platforms. XNNPACK is not intended for direct use by deep learning practitioners researchers; instead it provides low-level performance primitives for accelerating high-level machine learning frameworks, such as MediaPipe, TensorFlow Lite, and TensorFlow.js.
XNNPACK implements the following neural network operators:
All operators in XNNPACK support NHWC layout, but additionally allow custom stride along the Channel dimension. Thus, operators can consume a subset of channels in the input tensor, and produce a subset of channels in the output tensor, providing a zero-cost Channel Split and Channel Concatenation operations.
The table below presents single-threaded performance of XNNPACK library on two generations of MobileNet models and three generations of Pixel phones.
Model | Pixel, ms | Pixel 2, ms | Pixel 3a, ms |
---|---|---|---|
MobileNet v1 1.0X | 81 | 93 | 88 |
MobileNet v2 1.0X | 48 | 58 | 54 |
Benchmarked on October 9, 2019 with end2end_bench --benchmark_min_time=5
on an Android/ARM64 build (bazel build -c opt --config android_arm64 :end2end_bench
) and neural network models with randomized weights and inputs.
XNNPACK is a based on QNNPACK library. However, unlike QNNPACK, XNNPACK focuses entirely on floating-point operators, and its API is no longer compatible with QNNPACK.